Leak testing of heat exchanger tube bundles
Thread Content
Recently, I discussed with a maintenance company ways to simply test for leaks in heat exchanger tube bundles. One approach involves using a vacuum pump along with a plug: one end of the tube bundle is sealed while vacuum is created at the other end, and then a vacuum gauge is used to determine whether there are any damages to the tube bundle. Is this method reliable? What are its advantages and disadvantages?
Can we have a simple schematic diagram?
What was said upstairs is correct; generally, vacuum tests cannot achieve the pressures used in strength tests, and can only be used as a check for sealing integrity.
It is recommended to follow the specifications and standards for pressure vessels...
It is recommended to conduct a pressure test; once it passes, release some pressure to check for air leakage, and apply soapy water to the pipe ends to detect any leaks – this approach is more reliable.
This method is good; make sure the plug has a good seal
This won’t work; there are two types of internal leakage in heat exchangers: 1. Corrosion and perforation of the tube bundle. 2. Expansion joint between tube bundle and tube sheet ； Your method can only determine whether there is a leak in the first scenario; it cannot determine whether there is a leak in the second scenario.
This method is unreliable. 1. The sealing performance of the plunger cannot be guaranteed; whether the vacuum level is too high or too low, the plunger may leak air ； 2. I agree with what was said on floor 8: leaks in the heat exchange tube bundle can occur either in the tubes themselves or at the welding joints. By doing this, even if it is successful, it will only ensure that the tubes do not leak; it cannot guarantee that there are no leaks at the welding joints.
It’s better to follow the standard procedures and use ammonia leak testing or helium leak testing
Generally, water is used for pressure testing in the shell side; specialized fixtures are available, which makes it more reliable
Your method can only yield indirect evidence.